Global NEST Journal,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Nov. 25, 2024
<p><span
style="font-family:"Times
New
Roman","serif";font-size:12.0pt;line-height:150%;"
lang="EN-US">Water
quality
prediction
and
classification
plays
a
crucial
role
in
ecosystem
sustainability,
agriculture,
aquaculture
environmental
monitoring.
The
nonlinearity
nonstationarity
of
water
are
challenging
for
traditional
techniques
to
adequately
capture.
rapid
advancement
deep
learning
recent
decades
has
made
it
hot
topic
predicting
classification.
In
this
paper,
new
Optimization
driven
Deep
Differential
RecurFlowNet
(ODD-RecurFlowNet)
model
with
feature
selection
is
proposed
categorizing
the
quality.
Preprocessing
methods
utilized
evaluate
collected
data
predict
class
index.
Before
deploying
algorithm,
preprocessing
procedures
such
as
cleaning
robust
scalar
normalization
carried
out.
A
logistic
based
giant
armadillo
optimization
algorithm
(GArO)
used
optimal
selection.
Next,
index
predicted
using
global
attention
(GA)
model.
Subsequently,
differential
convolution
neural
network
(DDiff-CNN)
employed
different
levels
addition,
hyper-parameters
ODD-RecurFlowNet
tuned
crested
porcupine
(CPoOA).
For
simulation,
python
platform
standard
dataset
from
Kaggle
library
validate
experiment.
finding
shows
that
obtains
overall
accuracy
98.01%
RMSE
value
0.039.
Thus,
obtained
results
prove
superiority
existing
methods.</span></p>
PLoS ONE,
Journal Year:
2025,
Volume and Issue:
20(1), P. e0310218 - e0310218
Published: Jan. 24, 2025
Diabetes,
a
chronic
condition
affecting
millions
worldwide,
necessitates
early
intervention
to
prevent
severe
complications.
While
accurately
predicting
diabetes
onset
or
progression
remains
challenging
due
complex
and
imbalanced
datasets,
recent
advancements
in
machine
learning
offer
potential
solutions.
Traditional
prediction
models,
often
limited
by
default
parameters,
have
been
superseded
more
sophisticated
approaches.
Leveraging
Bayesian
optimization
fine-tune
XGBoost,
researchers
can
harness
the
power
of
data
analysis
improve
predictive
accuracy.
By
identifying
key
factors
influencing
risk,
personalized
prevention
strategies
be
developed,
ultimately
enhancing
patient
outcomes.
Successful
implementation
requires
meticulous
management,
stringent
ethical
considerations,
seamless
integration
into
healthcare
systems.
This
study
focused
on
optimizing
hyperparameters
an
XGBoost
ensemble
model
using
optimization.
Compared
grid
search
(accuracy:
97.24%,
F1-score:
95.72%,
MCC:
81.02%),
with
achieved
slightly
improved
performance
97.26%,
MCC:81.18%).
Although
improvements
observed
this
are
modest,
optimized
represents
promising
step
towards
revolutionizing
treatment.
approach
holds
significant
outcomes
for
individuals
at
risk
developing
diabetes.
Scientific Reports,
Journal Year:
2024,
Volume and Issue:
14(1)
Published: July 16, 2024
Abstract
In
regions
like
Oman,
which
are
characterized
by
aridity,
enhancing
the
water
quality
discharged
from
reservoirs
poses
considerable
challenges.
This
predicament
is
notably
pronounced
at
Wadi
Dayqah
Dam
(WDD),
where
meeting
demand
for
ample,
superior
downstream
proves
to
be
a
formidable
task.
Thus,
accurately
estimating
and
mapping
indicators
(WQIs)
paramount
sustainable
planning
of
inland
in
study
area.
Since
traditional
procedures
collect
data
time-consuming,
labor-intensive,
costly,
resources
management
has
shifted
gathering
field
measurement
utilizing
remote
sensing
(RS)
data.
WDD
been
threatened
various
driving
forces
recent
years,
such
as
contamination
different
sources,
sedimentation,
nutrient
runoff,
salinity
intrusion,
temperature
fluctuations,
microbial
contamination.
Therefore,
this
aimed
retrieve
map
WQIs,
namely
dissolved
oxygen
(DO)
chlorophyll-a
(Chl-a)
(WDD)
reservoir
Sentinel-2
(S2)
satellite
using
new
procedure
weighted
averaging,
Bayesian
Maximum
Entropy-based
Fusion
(BMEF).
To
do
so,
outputs
four
Machine
Learning
(ML)
algorithms,
Multilayer
Regression
(MLR),
Random
Forest
(RFR),
Support
Vector
(SVRs),
XGBoost,
were
combined
approach
together,
considering
uncertainty.
Water
samples
254
systematic
plots
obtained
(T),
electrical
conductivity
(EC),
(Chl-a),
pH,
oxidation–reduction
potential
(ORP),
WDD.
The
findings
indicated
that,
throughout
both
training
testing
phases,
BMEF
model
outperformed
individual
machine
learning
models.
Considering
Chl-a,
WQI,
R-squared,
evaluation
indices,
MLR,
SVR,
RFR,
XGBoost
6%,
9%,
2%,
7%,
respectively.
Furthermore,
results
significantly
enhanced
when
best
combination
spectral
bands
was
considered
estimate
specific
WQIs
instead
all
S2
input
variables
ML
algorithms.
